With more than one million victims and associated expenses of almost $127 billion each year, sexual violence (SV) is more costly than any other type of crime, including fatal crime and child abuse. In addition to societal and interpersonal costs, the impact on the individual can be high, including increased rates of posttraumatic stress disorder (PTSD), physical health problems, and suicidal threats and attempts. In short, SV is associated with significant public health burden. Despite the compelling data documenting the toll SV represents, important gaps in our understanding of the etiology of SV hamper our ability to develop and implement effective intervention and prevention strategies. Although perpetration likely emerges in adolescence, little is known about adolescent perpetrators - and this is especially true for females, lesbian, gay, bisexual and transgender (LGBT) youth, non-English speaking youth, and non-referred youth (i.e., youth who have not come into contact with the police). Indeed, of the handful of longitudinal studies focused on perpetrators, only two examine adolescents and only the Growing up with Media (GuwM) study includes female perpetrators. GuwM was designed to study the emergence of sexual violence in adolescence. Findings suggest that 15-16 years of age is a pivotal time period for the emergence of SV. Given the few number of youth who provided data when they were 15 years of age or younger, additional youth in the cohort are needed who age into this period of risk, to better understand factors that contribute to the etiology of SV. Findings also need to be extended into young adulthood so that we may better understand the developmental progression of sexual violence. The best available data are based upon justice records. Given that none of the perpetrators identified in GuwM reported criminal charges, there is a wide swath of youth not reflected in these crime data. To further our knowledge of how sexual violence emerges in adolescence and continues through young adulthood, we propose to continue following the GuwM cohort for another three years. To increase the number of perpetrators surveyed at younger ages when the behaviors are likely to emerge, we propose to also recruit a new cohort of 1600 youth 14-15 years of age that we will follow through the age of posited greatest risk. We aim to conduct an epidemiological study that will allow us to: Specific Aim 1: Investigate the developmental progression of onset and co-occurrence of different types of sexual violence perpetration. Specific Aim 2: Ascertain the degree of persistence, desistence and emergence of sexual violence perpetration in young adulthood depending on developmental patterns of sexual violence that emerge in adolescence. Specific Aim 3: Assess the predictive power of individual-, family-, and peer-level characteristics with respect to the time of onset of sexual violence perpetration and with respect to latent transitions of sexual violence perpetration profiles.